Resent-Date: Tue, 24 Aug 1999 10:58:15 +0200 (MET DST) Date: Tue, 24 Aug 1999 10:57:26 +0200 (MET DST) From: Roman Zippel X-Sender: zippel@zeus To: Linux/m68k Subject: 2.3.14 fix Resent-From: linux-m68k@phil.uni-sb.de Hi, The following patch is needed to do any user space debugging, I shouldn't have put ret_from_fork into the exit of do_trace. :) bye, Roman --- entry.S.org Fri Aug 20 20:46:56 1999 +++ entry.S Tue Aug 24 01:00:37 1999 @@ -72,6 +72,16 @@ pea SYMBOL_NAME(ret_from_exception) jmp SYMBOL_NAME(schedule) + | After a fork we jump here directly from resume, + | so that %d1 contains the previous task + | Theoretically only needed on SMP, but let's watch + | what happens in schedule_tail() in future... +ENTRY(ret_from_fork) + movel %d1,%sp@- + jsr SYMBOL_NAME(schedule_tail) + addql #4,%sp + jra SYMBOL_NAME(ret_from_exception) + badsys: movel #-ENOSYS,PT_D0(%sp) jra SYMBOL_NAME(ret_from_exception) @@ -88,16 +98,6 @@ subql #4,%sp | dummy return address SAVE_SWITCH_STACK jbsr SYMBOL_NAME(syscall_trace) - - | After a fork we jump here directly from resume, - | so that %d1 contains the previous task - | Theoretically only needed on SMP, but let's watch - | what happens in schedule_tail() in future... -ENTRY(ret_from_fork) - movel %d1,%sp@- - jsr SYMBOL_NAME(schedule_tail) - addql #4,%sp - jra SYMBOL_NAME(ret_from_exception) SYMBOL_NAME_LABEL(ret_from_signal) RESTORE_SWITCH_STACK